Update 第二阶段/Model_file/模型训练使用使用说明.md, 第二阶段/Model_file/main.py files

第二阶段/Model_file/main.py

@@ -11,9 +11,26 @@ from tensorflow.keras.optimizers import *
 from tensorflow.keras.datasets import *
 from pathlib import Path
 
+MANIFEST_DIR = sys.path[0] + "\\train.csv"
+Mode_DIR = sys.path[0] + "\\models"
+
+isExists=os.path.exists(Mode_DIR)
+
+# 判断结果
+
+if not isExists:
+
+    # 如果不存在则创建目录
+    # 创建目录操作函数
+    os.makedirs(Mode_DIR)
+    print(Mode_DIR + ' 创建成功')
+else:
+    # 如果目录存在则不创建，并提示目录已存在
+    print(Mode_DIR + ' 目录已存在')
+
+
 
 
-MANIFEST_DIR = sys.path[0] + "\\train.csv"
 
 Batch_size = 20
 Long = 792
@@ -58,25 +75,25 @@ def xs_gen(path=MANIFEST_DIR,batch_size = Batch_size,train=True,Lens=Lens,flag =
                 batch_y = np.array([convert2oneHot(label,10) for label in batch_list[:,-1]])
                 yield batch_x, batch_y
 
-TEST_MANIFEST_DIR = sys.path[0] + "\\test_data.csv"
+# TEST_MANIFEST_DIR = sys.path[0] + "\\test_data.csv"
 
-def ts_gen(path=TEST_MANIFEST_DIR,batch_size = Batch_size):
+# def ts_gen(path=TEST_MANIFEST_DIR,batch_size = Batch_size):
 
-    img_list = pd.read_csv(path)
+#     img_list = pd.read_csv(path)
 
-    img_list = np.array(img_list)[:Lens]
-    print("Found %s train items."%len(img_list))
-    print("list 1 is",img_list[0,-1])
-    steps = math.ceil(len(img_list) / batch_size)    # 确定每轮有多少个batch
+#     img_list = np.array(img_list)[:Lens]
+#     print("Found %s train items."%len(img_list))
+#     print("list 1 is",img_list[0,-1])
+#     steps = math.ceil(len(img_list) / batch_size)    # 确定每轮有多少个batch
 
-    for i in range(steps):
+#     for i in range(steps):
 
-        batch_list = img_list[i * batch_size : i * batch_size + batch_size]
-        #np.random.shuffle(batch_list)
-        batch_x = np.array([file for file in batch_list[:,1:]])
-        #batch_y = np.array([convert2oneHot(label,10) for label in batch_list[:,-1]])
+#         batch_list = img_list[i * batch_size : i * batch_size + batch_size]
+#         #np.random.shuffle(batch_list)
+#         batch_x = np.array([file for file in batch_list[:,1:]])
+#         #batch_y = np.array([convert2oneHot(label,10) for label in batch_list[:,-1]])
 
-        yield batch_x
+#         yield batch_x
 
 
 
@@ -124,7 +141,7 @@ if __name__ == "__main__":
 
         
         ckpt = tensorflow.keras.callbacks.ModelCheckpoint(
-            filepath= r"C:\Users\years\Desktop\learn_some\model\'best_model.{epoch:02d}-{val_loss:.4f}.h5",
+            filepath= Mode_DIR + "\\best_model.{epoch:02d}-{val_loss:.4f}.h5",
             monitor='val_loss', save_best_only=True,verbose=1)
 
         model = build_model()
@@ -143,7 +160,7 @@ if __name__ == "__main__":
                 callbacks=[ckpt],
                 )
 
-        keras_file = r"C:\Users\years\Desktop\learn_some\model\finishModel.h5"
+        keras_file = Mode_DIR + "\\finishModel.h5"
         model.save(keras_file, save_format="h5")
 
 
@@ -171,7 +188,7 @@ if __name__ == "__main__":
         
         tflite_model = converter.convert()
         
-        tflite_file = Path(r"C:/Users/years/Desktop/learn_some/model/finishModel.tflite")
+        tflite_file = Path(Mode_DIR + "\\finishModel.tflite")
         tflite_file.write_bytes(tflite_model)
         print("convert model to tflite done...")
     #     ######
@@ -210,7 +227,7 @@ if __name__ == "__main__":
 
 
 
-        keras_file_tflite = "C:/Users/years/Desktop/learn_some/model/finishModel.tflite"
+        keras_file_tflite = Mode_DIR + "\\finishModel.tflite"
         interpreter = tensorflow.lite.Interpreter(model_path=keras_file_tflite)
         interpreter.allocate_tensors()
         input_details = interpreter.get_input_details()

第二阶段/Model_file/模型训练使用使用说明.md

 环境说明：
 
+​	vscode
+
 ​	tensorflow 2.5.0 -  GPU
 
 ​	 CUDA Version: 11.4
 
-​	cuDNN:8.2.2
+​	 cuDNN:8.2.2
 
 请将训练数据与本程序放在同一目录下
 
 
 
-
-模型部分路径使用的绝对路径，使用时还请做出修改
-
 本程序适用于文件夹下的数据，如若使用其他数据请另作修改